Introduction

      Migration is one of the most widely observed phenomena in biology, occurring in all major branches of the animal kingdom (Dingle 2014). Migrations can be impressive, covering hundreds of kilometers and crossing through multiple biomes, making them incredibly taxing and dangerous for the migrating individual (Chapman et al. 2014). Therefore, the decision to migrate is dictated by a balance between the gain associated with a new habitat, such as enhanced feeding and access to breeding sites, and the potential for greater mortality and energy expenditure (Secor 2015; Bernatchez and Dodson 2011).
      In aquatic environments, fish that migrate across salinity boundaries to complete their life history are defined as diadromous. Anadromy, a form of diadromy, describes fish that spend the majority of their life in saltwater and return to freshwater to breed (Chapman et al. 2014; Myers 1949). These fishes undergo at least two major migrations in their life where they cross the salinity boundary, once traveling from freshwater to saltwater as larvae or juveniles, and a second time traveling back into the freshwater environment as adults to spawn. Anadromous fish are very important to aquatic communities because they forge an ecological significant link between marine and freshwater ecosystems (Secor 2015). With an extended range across two major ecosystems, changes in abundance or behavior of anadromous species can impact community structures on a much broader scale, compared to a resident species living in just one ecosystem (Holdo et al. 2011).
      Rainbow smelt (Osmerus mordax) are an anadromous species in decline. Historically, their geographic range extended from Chesapeake Bay to Labrador (Kendall 1962, Buckley 1989). However, over the past few centuries, this range has decreased significantly, currently being restricted to Massachusetts and northwards (Enterline et al. 2012). Populations of rainbow smelt in the southern extent of their range (Massachusetts, New Hampshire, and southern Maine) have been displaying indications of population stress: truncated age structures within the spawning run, high male to female ratios in some rivers, and a higher portion of age-1 spawners than historically observed (Enterline et al. 2012).These indicators have also been reflected in population abundance, with current harvests of rainbow smelt at historically low levels (Chase 2006). Rainbow smelt play an important role in supporting popular recreational fisheries in New Hampshire and Maine where they are caught by ice-fishing (December-March). They also contribute to important fisheries in coastal New England as a prey item for other recreational fish such as striped bass (Morone saxatilis), bluefish (Pomatomus salatrix), and Gulf of Maine Atlantic salmon (Salmo salar) which are listed under the Endangered Species Act (Collette and Klein-MacPhee 2002). In 2004, in response to declining populations of rainbow smelt and taking into consideration their importance in commercial and recreational fisheries, the National Oceanographic and Atmospheric Association’s (NOAA) National Marine Fisheries Service (NMFS) listed rainbow smelt as a species of concern in the Gulf of Maine (Chase 2006).
     As an ecologically and culturally significant species of concern, understanding key life history stages of the rainbow smelt is important to maintain long-term population stability. One emerging tool that has shown promise in further understanding nuances in fish movement is acoustic telemetry. Acoustic telemetry has been shown to be a powerful method to monitor the movement patterns of individual aquatic animals, including rainbow smelt (Hussey et al. 2015; Donaldson et al. 2014, Enterline et al. 2012). It can be used to quantify migratory behaviors and habitat use across freshwater, estuarine, and marine environments (Furey et al. 2016; Clark et al. 2016; Furey et al. 2013). Presence of individuals can be recorded along a known array, allowing for paths of movement and general duration of stay to be determined for an individual (Hussey et al. 2015). Acoustic telemetry does have limitations. For example, tagged individuals are not recaptured, which means data collection is limited by the battery life of the tag. Additionally, data collection relies on proximity of the tag to a receiver, meaning that if a tagged individual leaves the array, data collection will cease. These limitations mean that only a portion of an individual’s life can be monitored during each tagging period.
     A complementary technique to acoustic telemetry is otolith microchemistry, a method that can provide additional insight into individual habitat use over an entire life cycle. The chemical composition of otoliths, the ear bones found in teleost fish, can be used to identify natal origins, assess individual migratory behavior, quantify growth, and determine duration of residency in different salinity regimes (Walther 2019). Broad-scale migration patterns can be determined by looking at the uptake of elements into the otolith. Specifically, this method measures variation in elemental ratios which differ between fresh water and marine ecosystems, providing key insight into the environment an individual was in (Walther and Limburg 2012). This can be combined with age and growth analysis to determine duration of residency within each environment, allowing for lifelong migration patterns to be analyzed. Water samples can also be taken at sampling sites to provide an expected elemental composition of the water as a comparison for uptake into the otolith; this can aid in finer-scale differentiations, especially if there are differences in geological formations that may cause distinct elemental increases in the water (Walther et al. 2008).
      While past studies have illuminated general trends in smelt movement based on factors such as geographic latitude, river size, and distance of spawning grounds from marine water (Collette and Klein-MacPhee 2002), the specific timing in movement and key life history stages aligning with this movement have not been not well defined. Through utilizing acoustic telemetry, otolith microchemistry and additional environmental data, this study seeks to explore rainbow smelt movement within an estuarine environment. Specifically, we aim to determine smelt movement within the estuary, use of key habitats within that estuary, and the timelines and smelt life history stages associated with that movement.

Methods

Acoustic Telemetry
     A total of 44 adult rainbow smelt from the Great Bay Watershed were tagged with acoustic transmitters in April 2021 (Vemco V5, Bedford, Nova Scotia, Canada, 12.7-mm long, 5.7-mm tall, 5.8-mm wide, 180 kHz). Smelt were tagged in three different rivers within the system, the Oyster, Squamscott, and Bellamy Rivers, and were caught in collaboration with New Hampshire Fish and Game’s (NHFG) annual fyke net survey. No smelt were tagged from the Winnicut River due to the low population numbers at the time of tagging. Tags were programmed to transmit every 80-160 seconds and had an estimated battery life of 184 days. The population of tagged smelt was representative of the available populations during the time of capture. In preparation for tagging, fish were immersed in in an anesthetic bath of MS-222 buffered with sodium bicarbonate (100-mg MS-222: 200-mg sodium bicarbonate: 1-L river water) until they were unresponsive. Once unresponsive, fish were held on a custom-built tagging table with a v-shaped trough lined with damp paper towels and water was pumped continuously across the gills (via plastic tubing). Mass (g), total length (mm), and fork length (mm) were measured. A scalpel was used to make an incision (~6-mm) was along the ventral surface, starting just in front of the pelvic fins and moving towards the head. The acoustic tag was inserted into the body cavity and the incision was closed using two interrupted simple surgeon’s knots (size 5-0 monofilament suture with 26 mm 3/8 circle reverse-cutting needle, Ethicon, Somerville, NJ). Post-surgery, fish were held in aerated coolers to allow for recovery and observation. Once normal behavior was observed (regained mobility and equilibrium) fish were released at the site of capture.
     In March and April 2021, 18 VEMCO VR2W-180 kHz receivers were deployed throughout the system to listen for tagged rainbow smelt (Fig. 1). Receivers were strategically placed to monitor movement patterns within the tributaries, the Great Bay Estuary, and in and out of the estuary. Where possible, receivers were placed in geographical chokepoints or were placed on either side of the channel to increase the probability that a tagged fish would pass by within the proper range for data to be recorded. Each receiver was pulled approximately once per month until the end of October 2021 to offload data, check battery status, and perform general maintenance as needed before redeployment.

Map showing three different sampling methods (fyke netting, acoustic telemetry receivers, and water sampling) conducted within the Great Bay watershed to assess movement of adult and juvenile rainbow smelt

Map showing three different sampling methods (fyke netting, acoustic telemetry receivers, and water sampling) conducted within the Great Bay watershed to assess movement of adult and juvenile rainbow smelt

Otolith Microchemistry
     Adult and juvenile rainbow smelt were sampled for otolith microchemistry. Adult fish were obtained during NHFG’s 2021 creel and fyke net surveys. Creel surveys involve interviewing anglers and measuring their catch to monitor catch per unit effort and gain insight on the success of the recreational fishery. NHFG’s creel survey was performed on the Oyster and Winnicut Rivers before ice-off (February 2021). Smelt were collected from anglers on a volunteer basis; either the whole fish was collected, or the head was obtained so the angler could keep the body for consumption. Adult smelt obtained through fyke net surveys (March-April 2021) were collected from the Oyster, Squamscott, Winnicut, and Bellamy Rivers (Fig. 1).
     Juvenile rainbow smelt were sampled in collaboration with NHFG’s annual juvenile finfish beach seine survey (June -November 2021) at 11 sites in the Great Bay Estuary and leading out towards the Gulf of Maine. A bag seine with 6.4 mm mesh was used to sample at each site within two hours of low tide. All fish caught in the seine were identified to the lowest possible taxon, measured to the nearest mm (up to 25 individuals of each species per seine haul), and then enumerated. A subsample of rainbow smelt juveniles from each site were taken for analysis. Fish collected for otoliths, regardless of sampling method, were frozen for later otolith extraction and analysis. Total length (mm), fork length (mm), mass (g), and sex were recorded for all fish, regardless of sample method.
     Water samples for dissolved elemental analysis were also taken at each sampling site and at other geographically important sites within Great Bay Estuary (Fig. 1). Water samples are important for otolith validation as the chemical signatures recorded in otoliths should reflect ambient water composition at the time of deposition (Bath et al. 2000; Campana et al. 1995; B. Walther & Thorrold 2006). Therefore, elemental analysis of water at collection sites will help validate the interpretation of otolith compositions and provide site-specific information to aid in finer spatial distribution analysis (Bradbury et al. 2008; B. D. Walther et al. 2008). Surface water samples were collected and filtered using 0.45- and 0.2-μm hydrophilic polytetrafluorethylene (PTFE) luer-lock filters (Shiller, 2003). Water was filtered through the 0.45-μm filter and 0.2-μm filters in sequence to remove all particulates. After collection samples were stored in low-density polyethylene (LDPE) bottles where they were acidified to a pH < 2 with ultrapure nitric acid and refrigerated until analysis (Shiller, 2003). All water and otolith analysis will be performed at the University of Texas Jackson School of Geosciences (UTJSG) quadrupole ICP-MS lab. Results will further illuminate trends in time allocation of rainbow smelt in marine, estuarine, and freshwater environments throughout an individual’s life history.

Preliminary Findings

Acoustic Telemetry
      A total of 13903 detections were collected throughout the receiver array from April-October 2021. Of the 22 receivers, 16 obtained detections. Of those 16, the most detections were found along the transition from Great Bay to the Piscataqua River (Fig. 2). The migration of tagged rainbow smelt was categorized into three locations: river, Great Bay, and exit. Fish were assumed to have left the system if they reached the seaward-most receivers in the Piscataqua River. Thirty-four tagged fish (out of 44) reached these receivers and were defined as surviving the migration out of the system. Residency was calculated as the duration of time between tagging and last detection of fish that successfully left the system (average of 38.9 +/- 1.2 days and a range of 22.7-57.2 days).

Frequency of detections at acoustic receivers placed in Great Bay Estuary. Larger circles indicate a greater number of pings.

Frequency of detections at acoustic receivers placed in Great Bay Estuary. Larger circles indicate a greater number of pings.


      Preliminary analysis of movement patterns show that most tagged fish remained in rivers until mid-April with some fish moving between rivers. In general, smelt then moved to Great Bay where they remained until early-May. Around mid-May, fish moved down the Piscataqua towards the exit of the estuary. Once fish were in the Piscataqua River, they appeared to move quickly towards the exit of the system. Inferred movements (assuming linear path of movement between receivers) can be animated to visualize these movement patterns. Individual animations provide insight towards movement patterns in relation to tagging site and sex. As a preliminary look at these movements, three individuals from the three different tagging rivers were animated using gganimate in Rstudio (Fig 3-5).

Figure 3. Inferred movements of male rainbow smelt tagged in the Bellamy River


      The first animation shows a male rainbow smelt tagged in the Bellamy River (tag ID #53786) (Fig. 3). Though this fish was tagged in the Bellamy River, the first detection is from the Lamprey River. This is likely because most receivers were deployed simultaneously with tagging, with some receivers deployed after tagging events. Fish 53786 then moved up through Great Bay and back up the Bellamy River, where it remained for a few days before moving back to Great Bay. Finally, about a month after it was tagged, fish 53786 moved down the Piscataqua River, making its way to the exit in only a couple of days.

Figure 4. Inferred movements of female rainbow smelt tagged in the Squamscott River


     The second animation shows a female raindbow smelt tagged in the Squamscott River (tag ID #53774) (Fig. 4). Fish 53774 was also first detected in the Lamprey River. From there it moved to Great Bay, over to the Lamprey River, and back out to Great Bay over a few days. Fish 53774 then spent about a month in Great Bay moving back and forth from Great Bay proper towards the Piscataqua River. Finally, at the beginning of May, it traveled down the Pisactaqua River towards the exit in a span of two days.

Figure 5. Inferred movements of female rainbow smelt tagged in the Squamscott River


      The third animation shows a male rainbow smelt tagged in the Oyster River (tag ID #53801) (Fig. 5). This fish was first detected in the Oyster River and then it moved into Great Bay. Fish 53801 went back up the Oyster River once more before traveling around Great Bay for about a week. In mid-April, this fish moved part-way down the Piscataqua River before heading back up to Great Bay. At the end of April, it finally migrated towards the exit, making the whole trip down the Piscataqua River in one day.

Figure 6. Mean total length of rainbow smelt sampled for tagging verus otolith microchemistry analysis. Samples taken across four rivers (n= 7, 21, 18 for tagged fish from the Bellamy, Oyster, and Squamscott River, respectively; n= 11, 29, 31, 8 for otolith fish from the Bellamy, Oyster, Squamscott, and Winnicut River, respectively). Error bars represent standard error.

Figure 6. Mean total length of rainbow smelt sampled for tagging verus otolith microchemistry analysis. Samples taken across four rivers (n= 7, 21, 18 for tagged fish from the Bellamy, Oyster, and Squamscott River, respectively; n= 11, 29, 31, 8 for otolith fish from the Bellamy, Oyster, Squamscott, and Winnicut River, respectively). Error bars represent standard error.


Otolith Microchemistry
      Since life history stage can influence fish movement and success, a series of measurements were collected from all fish sampled across the different sampling methodologies and systems (sex, weight, total length). This was done to verify that fish of similar demographics were sampled regardless of sampling date, methodology, and river systems. Overall, the fish collected appear to be approximately the same average size with only slight variances among river, collection method, and sex (Fig. 6). These preliminary findings indicate that the collection goal was met even with the differences in collection time.

Conclusions and Next Steps


      At this stage of the project, all the acoustic telemetry data has been collected and we have a preliminary understanding of the outward migration of rainbow smelt from the Great Bay Estuary. Preliminary analysis shows that the majority of the smelt remained in the system for about a month before emigrating, indicating that the estuary is an important habitat for their life history. As part of this study, we were unable to observe the return migration. However, because receivers were active in the system until the beginning of November, and no rainbow smelt were observed to return, we can conclude that rainbow smelt return later than this date.
      More analyses need to be done on the acoustic telemetry data, such as quantifying survival rates, movement patterns, and duration of residency along different parameters including tagging site, sex, and total length. Animations provide insight towards these movement patterns but quantification is necessary for a comprehensive understanding.
      Much of the otolith microchemistry analysis is yet to be done. All otoliths are collected but many are not yet extracted. The otoliths need to be prepped for ICP-MS analysis by being embedded in epoxy, sectioned, and polished to reveal the core. While the acoustic telemetry data provides short-term, detailed movement patterns, the otoliths will provide long-term, rougher estimates of movement patterns. Therefore, once the otoliths are analyzed, they will compliment the acoustic telemetry data and hopefully provide information that was unable to be seen through tagging data, such as the timing of the return migration into the estuary and use of the estuary overwinter.


References

Bernatchez, L., and Dodson, J. J. 2011. Relationship between bioenergetics and behavior in anadromous fish migrations. Canadian Journal of Fisheries and Aquatic Sciences 44: 399-407.

Buckley, L.J. 1989. Species Profiles: Life Histories and Environmental Requirements of Coastal Fishes and Invertebrates (North Atlantic); Rainbow Smelt. US Fish and Wildlife Service Biol. Rep. 82(11.106). US Army Corps of Engineers, TR EL-82-4. 11pp.

Chapman, B.B., Hulthén, K., Wellenreuther, M., Hansson, L., Nilsson, J., and Brönmark, C. 2014. Patterns of animal migration. (Hansson and Åkesson, Ed.). Animal Movement Across Scales. Oxford University Press.
Chase, B.C. 2006. Rainbow smelt (Osmerus mordax) spawning habitat on the Gulf of Maine Coast of Massachusetts. Massachusetts Division of Marine Fisheries, Technical Report TR-30.

Clark, T.D., Furey, N.B., Rechisky, E.L., Gale, M.K., Jeffries, K.M., Porter, A.D., Casselman, M.T., Lotto, A.G., Patterson, D.A., Cooke, S.J. and Farrell, A.P. 2016. Tracking wild sockeye salmon smolts to the ocean reveals distinct regions of nocturnal movement and high mortality. Ecological applications 26(4): 959-978.

Collette, B.B., and G. Klein-MacPhee, G. 2002. Bigelow and Schroeder’s fishes of the Gulf of Maine. Washington and London: Smithsonian Institution Press.
Dingle, H. 2014. Migration: The biology of life on the move. Oxford University Press.

Donaldson, M.R., Hinch, S.G., Suski, C.D., Fisk, A.T., Heupel, M.R., & Cooke, S.J. 2014. Making connections in aquatic ecosystems with acoustic telemetry monitoring. Frontiers in Ecology and the Environment 12(10): 565-573.

Enterline, C.L., Chase, B.C., Carloni, J.M., and Mills, K.E. 2012. A Regional Conservation Plan for Anadromous Rainbow Smelt in the U.S. Gulf of Maine. 100 p. 
Furey, N. B., Dance, M. A., and Rooker, J. R. 2013. Fine-scale movements and habitat use of juvenile southern flounder Paralichthys lethostigma in an estuarine seascape. Journal of Fish Biology 82(5): 1469-1483.

Furey, N.B., Hinch, S.G., Bass, A.L., Middleton, C.T., Minke-Martin, V., and Lotto, A.G. 2016. Predator swamping reduces predation risk during nocturnal migration of juvenile salmon in a high-mortality landscape. Journal of Animal Ecology 85:948-959.

Holdo, R.M., Holt, R.D., Sinclair, A.R.E., Godley, B.J., and Thirgood, S. 2011. Migration impacts on communities and ecosystems: empirical evidence and theoretical insights. (Milner-Gulland, Fryxell, and Sinclair, Ed.). Animal Migration, a synthesis. Oxford University Press.

Hussey N.E., Kessel S.T., Aarestrup K., Cooke S., Cowley P., Fisk A., Harcourt R., Holland K., Iverson S., Kocik J., Mills Flemming J., and Whoriskey F. 2015. Aquatic animal telemetry: a panoramic window into the underwater world. Science 348:1255642.

Kendall, W.C. 1926. The Smelts. Bulletin of the United States Bureau of Fisheries 42.

Myers, G. S. 1949. Usage of anadromous, catadromous and allied terms for migratory fishes. Copeia 1949(2): 89-97.

Secor, D.H. 2015. Migration ecology of fishes. Johns Hopkins University Press, Baltimore.

Walther, B.D. 2019. The art of otolith chemistry: interpreting patterns by integrating perspectives. Marine and Freshwater Research. DOI: 10.1071/MF18270.

Walther, B.D., and Limburg, K.E. 2012. The use of otolith chemistry to characterize diadromous migrations. Journal of Fish Biology 81(2): 796-825.

Walther, B.D., Thorrold, S.R., and Olney, J.E. 2008. Geochemical signatured in otoliths record natal origins of American shad. Transactions of the American Fisheries Society 137(1): 57-69